JP7425165B1 - Adaptive equalization circuit, adaptive equalization method, and receiving device - Google Patents

Abstract

An object of the present invention is to detect equivalence convergence with high accuracy for a wide variety of modulation methods and to stably perform adaptive equalization processing. [Solution] When a polarization separation monitor 3 detects that a correlation value between a first polarization signal and a second polarization signal exceeds a predetermined value, a control circuit 4 causes a filter tap coefficient update circuit 2 to Re-run the initial convergence of tap coefficients. The polarization separation monitor 3 determines which quadrant of the four quadrants on the IQ plane the first polarization signal and the second polarization signal belong to, calculates a correlation value from the determination result, and calculates the correlation value. Compare the value with a predetermined value. If on-axis data located on the I-axis or Q-axis of the IQ plane exists in the first polarization signal and the second polarization signal, the sign determination circuit 7 extracts the on-axis data and extracts the on-axis data. The data on the positive side of the I-axis, the data on the negative side of the I-axis, the data on the positive side of the Q-axis, and the data on the negative side of the Q-axis are the values representing the four quadrants on the IQ plane, respectively. One of the values is associated with each other without duplication. [Selection diagram] Figure 2

Description

The present disclosure relates to an adaptive equalization circuit, an adaptive equalization method, and a receiving device that compensate for the characteristics of an optical transmission path in data communication.

In coherent optical communication, large-capacity transmission of several tens of Gbit/s or more is realized by compensating for distortion of a transmitted signal through digital signal processing on the receiving side. Digital signal processing mainly involves processes such as chromatic dispersion compensation, frequency control/phase adjustment, polarization multiplexing/demultiplexing, and polarization dispersion compensation.

Polarization multiplexing and polarization dispersion compensation processing is mainly performed by an adaptive equalizer. When an adaptive equalizer is implemented by digital signal processing, a digital filter is generally used. Distortion of the transmission signal can be compensated for by setting in the digital filter a filter tap coefficient that is calculated so that the distortion of the transmission signal is canceled out. The tap coefficients of the digital filter correspond to the impulse response of the filter characteristics. The tap coefficients are sequentially updated to adapt to temporally changing conditions, and the adaptive equalizer performs compensation that follows changes in the polarization state.

In addition, within the receiving optical module, the X-polarized signal (hereinafter referred to as A portion of the Y-polarized wave remains in the wave, and a portion of the X-polarized signal remains in the separated Y-polarized wave. In order to further separate X-polarized data and Y-polarized data, the digital filter in adaptive equalization includes two filters that output X-polarized data input in the X-polarized direction and Y-polarized direction, and a Y-polarized data filter. It is composed of a total of four filters, two filters that output in the X polarization direction and Y polarization direction in response to input polarization data.

The following sequential update algorithm is used to update the tap coefficients of these digital filters. RLS (Recursive Least-Squares), LMS (Least Mean Square), etc. are generally used as the sequential update algorithm. This involves inserting a known signal such as a training signal or pilot signal into an optical signal on the transmitting side, and minimizing the error between the transmitted known signal and the true value of this known signal (the value inserted on the transmitting side). This is an algorithm that updates the tap coefficients for each step size.

Furthermore, as a sequential update algorithm, a blind equalization method is recently used in which tap coefficients are obtained without using known signals. Blind equalization methods include Constant Modulus Algorithm (CMA) and Radius directed equalization (RDE), which extends CMA to multiple amplitude rings for use in QAM (Quadrature Amplitude Modulation). For example, see Patent Documents 1 and 2). In these, the tap coefficients are updated to minimize the error between the output of the digital filter and the desired value (the "desired value" can be easily estimated as the desired value of the amplitude in the case of a constant envelope). Ru. The tap coefficients are controlled and converged according to this algorithm.

However, equivalence convergence may occur in the above-described sequential update algorithm, particularly in the blind equalization method. Equivalence convergence refers to a state in which erroneous convergence occurs during the convergence process of an algorithm, and the filter outputs of X polarization data and Y polarization data become similar values. Mainly, both outputs may have values biased toward X polarization data or values biased toward Y polarization data. In this case, polarization separation between X polarization data and Y polarization data is not performed correctly. That is, equivalence convergence is an index for checking whether polarization separation is achieved when setting tap coefficients in adaptive equalization.

The occurrence of this equivalence convergence is determined from the correlation value of the output information of the adaptive equalization of the X polarization data and the Y polarization data, and indicates that polarization separation has not been achieved. In particular, there is a high probability that equivalence convergence will occur in blind equalization methods (CMA and RDE). When equivalence convergence occurs, the occurrence is detected and the sequential update algorithm is redone. At this time, the function that detects the occurrence of equivalence convergence is called an equivalence convergence monitor (MCC: Miss Capture Checker).

Japanese Patent Application Publication No. 2012-124782 Japanese Patent Application Publication No. 2021-190787

However, in conventional MCCs, depending on the modulation method, even if the polarization separation tap coefficients have converged correctly and polarization separation can be performed appropriately, equivalence convergence may be detected incorrectly. That is, depending on the modulation method, it may not be possible to accurately detect equivalence convergence. In this case, tap coefficients are updated frequently even when updating is not necessary, and the adaptive equalization process may become unstable.

The present disclosure has been made to solve the above-mentioned problems, and its purpose is to detect equivalence convergence with high accuracy for a wide variety of modulation methods and to stably perform adaptive equalization processing. The present invention provides an adaptive equalization circuit, an adaptive equalization method, and a receiving device that can perform the following steps.

The adaptive equalization circuit according to the present disclosure includes a digital filter that inputs a polarization-separated first polarization signal and a second polarization signal and performs further polarization separation processing; a filter tap coefficient updating circuit that updates tap coefficients of the digital filter accordingly; and comparing a correlation value between the first polarized signal and the second polarized signal output from the digital filter with a predetermined value. a polarization separation monitor; and a control circuit that causes the filter tap coefficient updating circuit to re-execute initial convergence of the tap coefficients when the polarization separation monitor detects that the correlation value exceeds the predetermined value. , the polarization separation monitor includes a sign determination circuit that determines which quadrant of four quadrants on the IQ plane the first polarization signal and the second polarization signal belong to; and the sign determination circuit. and a comparison circuit that compares the correlation value with the predetermined value. In the case of a transmission modulation method in which there is on-axis data located on the I-axis or Q-axis of , data on the negative side of the I axis, data on the positive side of the Q axis, and data on the negative side of the Q axis are each mapped to one of the values representing the four quadrants on the IQ plane without overlap. It is characterized by

According to the present disclosure, it is possible to detect equivalence convergence with high precision for a wide variety of modulation methods and to stably perform adaptive equalization processing.

FIG. 1 is a block diagram showing an optical communication system according to an embodiment. FIG. 2 is a diagram showing an adaptive equalization circuit according to an embodiment. FIG. 3 is a diagram showing a digital filter according to an embodiment. FIG. 3 is a diagram showing a polarization separation monitor according to an embodiment. FIG. 2 is a diagram showing a sign determination circuit according to an embodiment. 3 is a flowchart of the operation of the sign determination unit X. It is a flowchart of the operation of sign determination section Y. FIG. 3 is a diagram illustrating a first operation example of a sign determination circuit. FIG. 7 is a diagram illustrating a second operation example of the sign determination circuit. FIG. 7 is a diagram showing a third example of operation of the sign determination circuit. FIG. 7 is a diagram showing a third example of operation of the sign determination circuit. It is a figure which shows the comparative example of the code determination of 8QAM. FIG. 3 is a diagram showing a functional evaluation result in 8QAM of the adaptive equalization circuit according to the embodiment. FIG. 3 is a diagram showing a functional evaluation result in 8QAM of the adaptive equalization circuit according to the embodiment. FIG. 3 is a diagram showing a functional evaluation result in 8QAM of the adaptive equalization circuit according to the embodiment. FIG. 3 is a diagram showing a functional evaluation result in 8QAM of the adaptive equalization circuit according to the embodiment.

FIG. 1 is a block diagram showing an optical communication system according to an embodiment. The optical communication system includes a transmitter 100 and a receiver 200. The optical signal output from the transmitter 100 is transmitted to the receiver 200 through the optical fiber transmission line 300.

The transmitting device 100 includes a transmitting signal processing circuit 101 and a transmitting optical module 102. The receiving device 200 includes a receiving optical module 201 and a received signal processing circuit 202. The transmission signal processing circuit 101 performs predetermined processing on input data. Specifically, the transmission signal processing circuit 101 divides input data into horizontal polarization data X and vertical polarization data Y, and performs error correction encoding, band limit filtering, and modulation on each data. Performs processing such as mapping. The horizontally polarized data X and vertically polarized data Y that have been subjected to such processing are expressed as in-phase components and orthogonal components, respectively, and are output to the transmitting optical module 102. The in-phase component of the horizontal polarization data X is represented by X_I, and the orthogonal component is represented by X_Q. The in-phase component of the vertical polarization data Y is represented by Y_I, and the orthogonal component is represented by Y_Q. These data components are also indicated by the same symbols in receiving device 200, which will be described later.

Note that in this specification, "data" basically refers to "baseband data" and "signal" refers to "optical signal" or "high frequency signal." However, "X polarization" and "Y polarization" may indicate baseband data or high frequency signals.

The transmitting optical module 102 converts the horizontal polarization data X and the vertical polarization data Y into an X polarization signal and a Y polarization signal of optical signals, and combines and transmits the two polarization signals. This transmitting optical module 102 includes a signal light source 11 (signal LD), two 90° combiners 12 and 13, and a polarization combiner 14. The two 90° combiners 12 and 13 modulate the output light of the signal light source 11 with horizontal polarization data X and vertical polarization data Y, respectively, and convert it into an optical signal. The polarization combiner 14 combines the X polarization signal and the Y polarization signal converted into optical signals. The combined optical signal is transmitted to the receiving device 200 through the optical fiber transmission line 300.

In the receiving device 200, the receiving optical module 201 receives an optical signal, converts the received optical signal into an electrical signal, and outputs the electrical signal. The receiving optical module 201 includes a polarization separator 21, a local oscillation light source 22 (local LD), and two 90° hybrid circuits 23 and 24. The polarization separator 21 separates the optical signal into two orthogonal polarization components, an X polarization signal and a Y polarization signal.

The 90° hybrid circuits 23 and 24 combine each polarized signal of the optical signal output from the polarization separator 21 with the output light of the local oscillation light source 22, and further combine each polarized signal of the optical signal with an in-phase component I. Separate into orthogonal components Q. Note that the 90° hybrid circuits 23 and 24 have photoelectric converters (not shown). The photoelectric converter converts each component (I, Q) of the X-polarized signal and the Y-polarized signal, which are the optical signals output from the 90° hybrid circuits 23 and 24, into an electrical signal, and converts this electrical signal into an X-polarized signal. It outputs as wave data (X_I, X_Q) and Y polarization data (Y_I, Y_Q). Hereinafter, the X polarization data and the Y polarization data will be referred to as received signals. "Data" here indicates an analog baseband signal. Note that the above configuration for obtaining X polarization data and Y polarization data is an example and is not limited to the above configuration.

The received signal processing circuit 202 includes an AD converter 25, a chromatic dispersion compensation circuit 26, an adaptive equalization circuit 27, and a decoding circuit 28. The AD converter 25 converts the electrical signal output from the receiving optical module 201 into a digital signal. When an optical signal propagates through the optical fiber transmission line 300, the signal waveform is distorted due to chromatic dispersion. The chromatic dispersion compensation circuit 26 estimates the magnitude of distortion of the received signal from the digital signal output from the AD converter 25, and compensates for the distortion due to chromatic dispersion of the digital signal.

In addition, the polarization mode of the optical fiber transmission line 300 is Dispersion effects cause polarization fluctuations, which distort the signal waveform. The adaptive equalization circuit 27 performs equalization processing to compensate for distortion of the output signal of the chromatic dispersion compensation circuit 26 due to polarization fluctuations. Note that polarization separation is first performed in the receiving optical module 201, but the adaptive equalization circuit 27 further processes the polarization separation in a complete direction. The decoding circuit 28 decodes the received signal output from the adaptive equalization circuit 27 and reproduces the original data (ie, the input data of the transmitted signal processing circuit 101).

Note that until the data is decoded by the decoding circuit 28, the input/output of each processing circuit of the received signal processing circuit 202 is the in-phase component X_I and quadrature component X_Q of the X polarization data, and the in-phase component Y_I of the Y polarization data. and an orthogonal component Y_Q. In each processing circuit, the X polarization data and the Y polarization data are generally processed while being separated into in-phase components and orthogonal components, respectively. Although the X polarization data (X_I, X_Q) and the Y polarization data (Y_I, Y_Q) are coordinate data, the decoding circuit 28 converts them into logical data of "1" and "0".

FIG. 2 is a diagram showing an adaptive equalization circuit according to an embodiment. The adaptive equalization circuit 27 includes a digital filter 1 , a filter tap coefficient update circuit 2 , a polarization separation monitor 3 , and a control circuit 4 .

The digital filter 1 inputs the first polarization signal and the second polarization signal that have been polarized by the wavelength dispersion compensation circuit 26, and performs further polarization separation processing to compensate for distortion and the like. The compensated result is supplied to the filter tap coefficient updating circuit 2. Signals received through the optical fiber transmission line 300 are generally affected by variations in the polarization state of the optical fiber transmission line 300. Therefore, the input signal from the chromatic dispersion compensation circuit 26 is also affected by the change in the polarization state. The filter tap coefficient update circuit 2 adaptively updates the tap coefficients of the digital filter 1 according to changes in the polarization state using a sequential update algorithm. The updated tap coefficients are set in the digital filter 1. In the sequential update algorithm, the tap coefficients are sequentially updated and converged to a predetermined value so that the output of the digital filter 1 becomes the original value.

At this time, the input signals to the digital filter 1 are both X polarization data and Y polarization data. While the filter tap coefficients are being updated in the filter tap coefficient updating circuit 2, the output of the digital filter 1 is also supplied to the polarization separation monitor 3. This polarization separation monitor 3 is called an equivalence convergence monitor (MCC). The polarization separation monitor 3 always calculates the correlation value between the X polarization data and the Y polarization data output from the digital filter 1, compares the correlation value with a predetermined value, and separates the X polarization data and the Y polarization data. monitor whether the data is properly polarized. If the X polarization data and the Y polarization data are not properly polarized, a correlation greater than a threshold value will occur between the X polarization data and the Y polarization data. The polarization separation monitor 3 determines failure of polarization separation based on this correlation. When it is determined that the polarization separation has failed, the control circuit 4 causes the filter tap coefficient updating circuit 2 to re-execute the initial convergence of the tap coefficients. Initial convergence means monitoring the control circuit to determine whether polarization separation has been achieved, regenerating the Wiener filter using the least squares method, and regenerating the Wiener filter from the tap coefficients in the impulse state (only the center tap is set to 1). Execute a sequential update algorithm (CMA or RDE) or perform some processing from the current state of the taps, for example, keep only the taps on the polarization side with large tap coefficient amplitudes and impulse the tap coefficients on the other polarization side. Indicates, for example, re-executing from the tap coefficient of the state.

However, in reality, a frequency error compensation circuit and a carrier phase recovery circuit, which are not shown in the figure, are provided on the output side of the digital filter 1. Synchronization with the frequency and phase of the carrier is performed by these circuits, and X polarization data and Y polarization data after carrier synchronization are supplied to the polarization separation monitor 3 and the decoding circuit 28.

Note that the monitoring operation of the polarization separation monitor 3 may cause erroneous detection depending on the modulation method. Generally, in the case of modulation methods such as QPSK and 16QAM, monitor malfunctions are rare. However, the monitor may malfunction especially with modulation schemes such as 8QAM that include a signal in which transmitting side data is mapped to a point where the value of the in-phase component or quadrature component is around 0. If the monitor malfunctions, unnecessary tap coefficient updates will be repeated, making the adaptive equalization operation unstable.

In contrast, the polarization separation monitor 3 according to the present embodiment can reduce malfunctions of the monitor function in any modulation method. Furthermore, the adaptive equalization circuit 27 using the polarization separation monitor 3 can appropriately determine polarization separation, constantly update filter tap coefficients appropriately, and perform stable adaptive equalization operations. The output of the digital filter 1 whose tap coefficients have been appropriately updated in this way is supplied to the decoding circuit 28 shown in FIG. 2 as a compensated received signal.

FIG. 3 is a diagram showing a digital filter according to an embodiment. This digital filter 1 is an example composed of an FIR filter. However, the digital filter 1 is not limited to this configuration, and may have any configuration that can cause erroneous convergence in the case of a sequential update algorithm in which the tap coefficients of the filter are determined by the convergence operation of the convergence algorithm.

Examples of sequential update algorithms include blind equalization methods such as CMA (Constant Modulus Algorithm) and RDE (Radius Directed Equalization). These update the tap coefficients to minimize the error between the output of digital filter 1 and the value that should originally exist (the "ideal value" can be easily estimated as the desired value of the amplitude in the case of a constant envelope). do. Other examples include RLS (Recursive Least-Squares) and LMS (Least Mean Square). These methods insert a known signal such as a training signal or pilot signal into an optical signal on the transmitting side, and minimize the error between the transmitted known signal and the true value of this known signal (value set on the transmitting side). The tap coefficients are updated for each step size.

The digital filter 1 includes FIR (Finite Impulse Response) filters FIR_A, FIR_B, FIR_C, and FIR_D configured in a butterfly type. Each FIR filter comprises N taps. However, the number of taps of the FIR filters may be different from each other. FIR_A is a filter for X polarization data. FIR_B is a filter for the influence of Y polarization data on X polarization data. FIR_C is a filter for the influence of X polarization data on Y polarization data. FIR_D is a filter for Y polarization data.

Digital filter 1 uses the addition value of the filtering result of FIR_A for X polarization data and the filtering result of FIR_B for Y polarization data as a compensation output for X polarization data, and combines the filtering result of FIR_C for X polarization data and The sum of the wave data and the FIR_D filtering result is set as the compensation output of the Y polarization data. These further ensure polarization separation.

Digital filter 1 uses the addition value of the filtering result of FIR_A for X polarization data and the filtering result of FIR_B for Y polarization data as a compensation output for X polarization data, and combines the filtering result of FIR_C for X polarization data and The sum of the wave data and the FIR_D filtering result is set as the compensation output of the Y polarization data. These further ensure polarization separation.

Further, the filter tap coefficients of the digital filter 1 are determined and set by the filter tap coefficient updating circuit 2. At that time, the tap coefficients of FIR_A, FIR_B, FIR_C, and FIR_D are expressed by the following equations.
WXX(n+1)=WXX(n)+μeX(n)Xout(n)・Xin*(n)
WYX(n+1)=WYX(n)+μeX(n)Xout(n)・Yin*(n)
WXY(n+1)=WXY(n)+μeY(n)Yout(n)・Xin*(n)
WYY(n+1)=WYY(n)+μeY(n) Yout(n)・Yin*(n)
Here, n is a value indicating the update order in the sequential update algorithm. Tap coefficient WXX(n) indicates a group of tap coefficients of FIR_A in the case of update order n. Tap coefficient WYX(n) indicates a group of tap coefficients of FIR_B in the case of update order n. Tap coefficient WXY(n) indicates a group of tap coefficients of FIR_C in the case of update order n. Tap coefficient WYY(n) indicates a group of tap coefficients of FIR_D in the case of update order n. μ indicates the step size of the update algorithm. eX(n) indicates the error between the filter output of the X polarization data and the desired value. eY(n) indicates the error between the Y polarization data and the desired value in the filter output. Xout(n) indicates the filter output in the X polarization data. Xin(n) indicates the filter input in the X polarization data. Yout(n) indicates the filter output in Y polarization data. Yin(n) indicates the filter input in Y polarization data. * indicates conjugation or complex conjugation. Note that the data and tap coefficients are represented by complex numbers.

According to the above-described sequential update algorithm, the tap coefficients are sequentially updated in the update order n, and the tap coefficients finally converge. The convergence condition is determined by the number of times n is updated or the error between the filter output and the desired value. Note that the above formula is an example of a formula representing a sequential update algorithm, and the formula representing a sequential update algorithm is not limited to the above.
Although the above example shows the case where the tap coefficients are updated after convergence, it is also possible to update the tap coefficients sequentially during convergence. That is, even if the convergence condition is not achieved, a method may be adopted in which the tap coefficients are sequentially updated using tap coefficients calculated so as to minimize the difference between the output of the digital filter and the desired value in a state where the polarization state fluctuates.

FIG. 4 is a diagram showing a polarization separation monitor according to an embodiment. In FIG. 4, a frequency error compensation circuit 5 and a carrier phase recovery circuit 6, which are not shown in FIG. 2, are connected between the digital filter 1 and the decoding circuit 28. The frequency error compensation circuit 5 is a circuit that makes the frequency error between the transmission carrier and the receiver carrier substantially zero. The carrier phase recovery circuit 6 is a circuit that makes the phase error between these carriers substantially zero. By the frequency error compensation circuit 5 and the carrier phase recovery circuit 6, the X polarization data and the Y polarization data from the digital filter 1 can be stably displayed on the IQ plane without phase rotation or phase shift. Thereby, the X polarization data and Y polarization data from the digital filter 1 can be accurately processed by the polarization separation monitor 3 and the decoding circuit 28. Note that if the phase rotation and phase shift can be removed by other methods, the frequency error compensation circuit 5 and the carrier phase recovery circuit 6 are not essential to the adaptive equalization circuit 27.

The polarization separation monitor 3 includes a sign determination circuit 7, a correlation calculation circuit 8, and a comparison circuit 9. The sign determination circuit 7 determines the sign of the outputs Xout and Yout of the digital filter 1 supplied via the frequency error compensation circuit 5 and the carrier phase recovery circuit 6. Here, consider a case where the outputs Xout and Yout of the digital filter 1 are shown on the IQ plane. Specifically, Xout has an I-axis component X_I and a Q-axis component X_Q, and Yout has an I-axis component Y_I and a Q-axis component Y_Q. The sign determination circuit 7 determines which quadrant of the four quadrants on the IQ plane the received X-polarized data and Y-polarized data belong to on a bit-by-bit or symbol-by-symbol basis. Note that in the case of multilevel modulation such as QPSK or 16QAM, the symbol indicates a unit of change in phase or amplitude. In QPSK, 2 bits are one symbol, and in 16QAM, 4 bits are one symbol.

The correlation calculation circuit 8 calculates the correlation value between the X polarization data and the Y polarization data from the determination result of the sign determination circuit 7. The closer the X polarization data and the Y polarization data are to the same, the higher the correlation value becomes. The comparison circuit 9 compares the correlation value calculated by the correlation calculation circuit 8 with a predetermined value, and transmits the comparison result to the control circuit 4.

If the correlation value exceeds a predetermined value, it is assumed that there is a correlation between the X polarization data and the Y polarization data, and it is determined that the tap coefficients have not converged appropriately. Therefore, when the polarization separation monitor 3 detects that the correlation value exceeds a predetermined value, the control circuit 4 causes the filter tap coefficient updating circuit 2 to re-execute the initial convergence of the tap coefficients. On the other hand, if the correlation value from the correlation calculation circuit 8 is smaller than the predetermined value, the control circuit 4 considers that there is no correlation between the X polarization data and the Y polarization data, and the tap coefficients are appropriately converged. It is determined that the In that case, the control circuit 4 does not instruct the filter tap coefficient updating circuit 2 to re-execute the initial convergence of the tap coefficients, and the digital filter 1 continues to use the tap coefficients at that time.

FIG. 5 is a diagram showing a sign determination circuit according to an embodiment. The frequency error compensation circuit 5 and carrier phase recovery circuit 6 are omitted. The sign determination circuit 7 includes sign determination sections X, Y and averaging circuits A, B, C, and D.

The sign determination section do. The values indicating the four quadrants (1st quadrant, 2nd quadrant, 3rd quadrant, 4th quadrant) on the IQ plane are (+1, +1), (-1, +1), (-1, -1), respectively. (+1, -1). When the X polarization data belongs to the first quadrant, the sign is A=+1, B=+1, when it belongs to the second quadrant, the sign is A=-1, B=+1, and when it belongs to the third quadrant, the sign is A=+1, B=+1. A=-1, B=-1, and when belonging to the fourth quadrant, the signs are A=+1, B=-1.

The sign determination unit Y also determines to which quadrant of the four quadrants on the IQ plane the Y polarization data belongs based on the I-axis component Y_I and Q-axis component Y_Q of the Y-polarization data from the digital filter 1. do. The values indicating the four quadrants (1st quadrant, 2nd quadrant, 3rd quadrant, 4th quadrant) on the IQ plane are (+1, +1), (-1, +1), (-1, -1), respectively. (+1, -1). When Y polarization data belongs to the first quadrant, the sign is C=+1, D=+1, when it belongs to the second quadrant, the sign is C=-1, D=+1, and when it belongs to the third quadrant, the sign is C=+1, D=+1. is C=-1, D=-1, and when belonging to the fourth quadrant, the codes are C=+1 and D=-1.

In the case of on-axis data in which the X polarization data or Y polarization data is on the I axis or the Q axis, the quadrant to which it belongs is not clear. Therefore, the sign determination circuit 7 extracts the on-axis data, and among the on-axis data, data on the positive side on the I axis, data on the negative side on the I axis, data on the positive side on the Q axis, and data on the positive side on the Q axis. The data on the negative side of is associated with one of the values indicating the four quadrants on the IQ plane without overlap, and a code indicating the associated quadrant is output. Details will be discussed later.

The averaging circuit A accumulates the code A determined by the code determination unit X a predetermined number of times and outputs the accumulated code A as an averaged value. The averaging circuit B accumulates the code B determined by the sign determining section X a predetermined number of times and outputs the accumulated code B as an averaged value. The averaging circuit C accumulates the code C determined by the sign determination unit Y a predetermined number of times and outputs it as an averaged value C. The averaging circuit D accumulates the code D determined by the code determination unit Y a predetermined number of times and outputs it as an averaged value D.

Using averaging A, averaging B, averaging C, and averaging D, averaged data XA of X polarization and averaged data YA of Y polarization are expressed as follows. Note that j is an imaginary unit.
XA=averaging A+j averaging B
YA=averaging C+j averaging D

Subsequently, the averaged data XA of the X polarization and the averaged data YA of the Y polarization outputted from the sign determination circuit 7 are supplied to the correlation calculation circuit 8 . The correlation calculation circuit 8 calculates the correlation value between the averaged data XA of the X polarization and the averaged data YA of the Y polarization using the following formula. "*" indicates complex multiplication.
Correlation value = Σ(XA*YA)/√(ΣXA ² *ΣYA ² )

If the X polarization data X and the Y polarization data Y have the same value in the output of the digital filter 1 (if there is a correlation), the averaged data XA of the X polarization and the averaged data YA of the Y polarization are the same. Therefore, the above-mentioned correlation value becomes extremely close to 1. Note that when the signs of XA and XB become the same, the correlation value does not approach 0 but increases. On the other hand, if the X polarization data X and the Y polarization data Y are almost randomly different (= there is no correlation), then the values of the averaged data There is a probability of 1/2 that the signs will be different. As a result, the sign of the numerator of the correlation value, XA*YA, becomes positive or negative, and the cumulative value approaches zero.

Note that the cumulative numbers in the numerator and denominator of the correlation value equation are different from the predetermined number of times when averaging is performed in the sign determination circuit 7. For example, the averaging of the sign determination circuit 7 can be performed every 16 symbols, and the correlation value equation can be accumulated 512 times. In this case, a correlation value is obtained every 16×512=8192 symbols.

Further, the calculation of the correlation value by the correlation calculation circuit 8 is not limited to the above method. If it is possible to calculate the correlation value between the X polarization data X and the Y polarization data Y, which are the outputs of the digital filter, based on the averaged data XA of the X polarization and the averaged data YA of the Y polarization from the sign determination circuit 7. , any arithmetic expression can be applied to the adaptive equalization circuit 27 of this embodiment.

FIG. 6 is a flowchart of the operation of the sign determination section X. The steps performed by the code determination unit X differ depending on the modulation method.

Step S0: When a part of the data is mapped on the I-axis or the Q-axis on the IQ plane, such as when the modulation method is 8QAM, the process continues to step S1. On the other hand, if all data is not mapped onto the I-axis or Q-axis in the IQ plane, such as when the modulation method is QPSK or 16QAM, steps S1 to S3 are skipped and the process proceeds to step S4.

Step S1: When the I-axis component X_I and the Q-axis component X_Q of the X-polarized wave data are input from the digital filter 1, the power of the X-polarized wave data is estimated based on the following formula.
Power of X polarization data = absolute value of X_I + absolute value of X_Q

Although power is generally expressed as the square of a signal, it can also be expressed as an index of power using the above equation. The above formula allows calculations using only addition without using multiplication, so it is easier to obtain the power index. In modulation methods that mainly change phase and amplitude, the amplitudes that data can take are generally divided into several groups. In particular, the amplitudes that can be taken by the data of the modulation method used in the blind equalization method are narrowed down to several types. In the case of QPSK, there is one type of amplitude, in the case of 8QAM, there are two types of amplitude, and in the case of 16QAM, there are three types of amplitude. The power calculation in this step only determines which amplitude group the data belongs to, so there is no need to accurately measure the power value.

Step S2: Based on the magnitude of the power obtained in step S1, it is determined to which amplitude group the data belongs. For example, if there are two types of amplitude that the data can take, inner shell and outer shell, if the power of

Step S3: If it is determined in step S2 that it is an inner shell, the symbols A and B are determined according to the following conditions. Here, a case is shown in which inner shell data is mapped on the I-axis or Q-axis, and outer shell data is mapped within four quadrants of the IQ plane. Note that if the data of the inner shell is mapped within a quadrant on the IQ plane and the data of the outer shell is mapped on the I axis or the Q axis, step S3 is executed in the case of the outer shell, and step S3 is executed in the case of the inner shell. Step S4 is executed.

If the absolute value of X_I≧the absolute value of X_Q, if X_I≧0, the fourth quadrant is (A, B)=(+1,-1), and if , +1). If the absolute value of X_I<the absolute value of -1).

Step S4: If it is determined in step S2 that it is an outer shell, the symbols A and B are determined according to the following conditions. In the case of X_I≧0, if X_Q≧0, the first quadrant is determined to be (A, B)=(+1, +1), and if X_Q<0, the fourth quadrant is determined to be (A, B)=(+1, −1). If X_I<0, if X_Q≧0, determine the second quadrant as (A, B) = (-1, +1), and if X_Q<0, determine the third quadrant as (A, B) = (-1, -1). do.

Step S5: Output the codes A and B obtained in step S3 or step S4. Note that it is also possible to make the following determination in step S3. If the absolute value of X_I ≧ the absolute value of X_Q, if X_I ≧ 0, the first quadrant is (A, B) = (+1, +1), and if -1). If the absolute value of X_I<the absolute value of -1).

In the above example, the data can take on two types of amplitudes: the inner shell and the outer shell, but it can also be applied to a case where the data can take on three or more types of amplitudes. In that case, it is sufficient to determine which amplitude group the group belongs to using a similar power calculation method. After determining the amplitude group, quadrants are assigned to on-axis data located on the I-axis or Q-axis in step S3. Furthermore, for data that is originally mapped within a quadrant, the quadrant is determined by a hard decision in step S4.

Note that the correspondence between the on-axis data located on the I-axis or the Q-axis and the quadrants is not limited to the above two types. Different data may be assigned to any quadrant as long as they are not associated with the same quadrant. For example, (I, Q) = {(1, 0), (-1, 0), (0, 1), (0, -1)}, (A, B) = {(-1, +1) , (+1, -1), (-1, -1), (+1, +1)}, (A, B) = {(+1, -1), (-1, +1), (-1, -1 ), (+1, +1)}, or (A, B) = {(+1, +1), (-1, +1), (-1, -1), (+1, -1)}, etc. It is possible. As mentioned above, data on the positive side of the I axis of X polarization data and Y polarization data, data on the negative side of the I axis, data on the positive side of the Q axis, data on the negative side of the Q axis By associating each of them with the quadrants without overlap, the correlation between the X polarization data and the Y polarization data can be efficiently detected. Such arbitrary correspondence also applies to the sign determination unit Y described below.

FIG. 7 is a flowchart of the operation of the sign determination unit Y. The operation of the sign determination section Y is similar to the operation of the sign determination section X shown in FIG.

Step S0: When a part of the data is mapped on the I-axis or the Q-axis on the IQ plane, such as when the modulation method is 8QAM, the process continues to step S1. On the other hand, if all data is not mapped on the I-axis or Q-axis in the IQ plane, such as when the modulation method is QPSK or 16QAM, steps S1 to S3 are skipped and the process proceeds to step S4.

Step S1: When the I-axis component Y_I and Q-axis component Y_Q of Y-polarized data are input from the digital filter 1, the power of the Y-polarized data is estimated based on the following.
Power of Y polarization data = Absolute value of Y_I + Absolute value of Y_Q Note that power calculation only determines which amplitude group the data belongs to, so it can be compared using the index in the above formula, and accurate power No value measurement is required.

Step S2: Based on the magnitude of the power obtained in step S1, it is determined to which amplitude group the Y polarization data belongs. For example, if the Y polarization data has two types of amplitudes, inner shell and outer shell, if the Y power≧set threshold value, it is determined to be outer shell, and otherwise it is determined to be inner shell.

Step S3: If it is determined in step S2 that it is an inner shell, the codes C and D are determined according to the following conditions. Here, a case is shown in which inner shell data is mapped on the I-axis or Q-axis, and outer shell data is mapped within four quadrants of the IQ plane. Note that if the data of the inner shell is mapped within a quadrant on the IQ plane and the data of the outer shell is mapped on the I axis or the Q axis, step S3 is executed in the case of the outer shell, and step S3 is executed in the case of the inner shell. Step S4 is executed.

If the absolute value of Y_I ≧ the absolute value of Y_Q, if Y_I ≧ 0, the fourth quadrant is (C, D) = (+1, -1), and if Y_I < 0, the second quadrant is (C, D) = (-1). , +1). When the absolute value of Y_I<the absolute value of Y_Q, if Y_Q≧0, the first quadrant is (C, D)=(+1,+1), and if Y_Q<0, the third quadrant is (C, D)=(−1, -1).

Step S4: If it is determined in step S2 that it is an outer shell, the codes C and D are determined according to the following conditions. In the case of Y_I≧0, if Y_Q≧0, the first quadrant is determined to be (C, D)=(+1, +1), and if Y_Q<0, the fourth quadrant is determined to be (C, D)=(+1, −1). If Y_I<0, if Y_Q≧0, the second quadrant is determined as (C, D) = (-1, +1), and if Y_Q<0, the third quadrant is determined as (C, D) = (-1, -1). do.

Step S5: Output the codes C and D obtained in step S3 or step S4. Note that it is also possible to make the following determination in step S3. If the absolute value of Y_I≧the absolute value of Y_Q, if Y_I≧0, the first quadrant is (C, D)=(+1,+1), and if Y_I<0, the third quadrant is (C, D)=(−1, -1). When the absolute value of Y_I<the absolute value of Y_Q, if Y_Q≧0, the second quadrant is (C, D) = (-1, +1), and if Y_Q<0, the fourth quadrant is (C, D) = (+1, -1).

Note that the processing of the sign determination section Y can also be applied to the case where there are three or more types of amplitudes that data can take, similarly to the sign determination section X. Furthermore, the mapping of on-axis data located on the I-axis or Q-axis to quadrants is not limited to the above two types, and if different data is not mapped to the same quadrant, it can be assigned to which quadrant. Good too.

FIG. 8 is a diagram showing a first example of operation of the sign determination circuit. The first operation example is when the modulation signal is QPSK. When the modulation signal is QPSK, there is no on-axis data located on the I-axis or the Q-axis, so step S4 is executed after step S0 in the flowcharts of FIGS. 6 and 7.

In step S4, the sign determination unit It is determined as (+1, -1). In the case of X_I<0, the sign determination unit , -1). If Y_I≧0, the sign determination unit Y determines (C, D)=(+1,+1) as the first quadrant if Y_Q≧0, and (C, D)=(+1,−1) as the fourth quadrant if Y_Q<0. 1). In the case of Y_I<0, the sign determination unit Y sets (C, D)=(-1, +1) as the second quadrant if Y_Q≧0, and (C, D)=(-1) as the third quadrant if Y_Q<0. , -1).

As a result of the above, (A, B), (C, D) are the values indicating the first quadrant, the second quadrant, the third quadrant, and the fourth quadrant of QPSK on the receiving side. This corresponds to the result determined by hard decision. For example, the hard decision results for values indicating the first quadrant are A=+1, B=+1, C=+1, and D=+1. The hard decision results for the values indicating the second quadrant are A=-1, B=+1, C=-1, and D=+1. The hard decision results for the values indicating the third quadrant are A=-1, B=-1, C=-1, and D=-1. The hard decision results for the values indicating the fourth quadrant are A=+1, B=-1, C=+1, and D=-1.

FIG. 9 is a diagram showing a second operation example of the sign determination circuit. The second operation example is a case where the modulation signal is 16QAM. When the modulation signal is 16QAM, there is no on-axis data located on the I-axis or the Q-axis, so step S4 is executed after step S0 in the flowcharts of FIGS. 6 and 7.

In step S4, the sign determination unit It is determined as (+1, -1). In the case of X_I<0, the sign determination unit , -1). If Y_I≧0, the sign determination unit Y determines (C, D)=(+1,+1) as the first quadrant if Y_Q≧0, and (C, D)=(+1,−1) as the fourth quadrant if Y_Q<0. 1). In the case of Y_I<0, the sign determination unit Y sets (C, D)=(-1, +1) as the second quadrant if Y_Q≧0, and (C, D)=(-1) as the third quadrant if Y_Q<0. , -1).

The above results (A, B) and (C, D) correspond to the results of determining the coordinates of all 16 16QAM data mapped on the IQ plane by hard decision on the receiving side. For example, the hard decision results for values indicating the first quadrant are A=+1, B=+1, C=+1, and D=+1. The hard decision results for the values indicating the second quadrant are A=-1, B=+1, C=-1, and D=+1. The hard decision results for the values indicating the third quadrant are A=-1, B=-1, C=-1, and D=-1. The hard decision results for the values indicating the fourth quadrant are A=+1, B=-1, C=+1, and D=-1.

As described above, for the data in the four quadrants on the IQ plane, A, B, C, and D are determined by hard decisions using the I and Q axes. Therefore, in a transmission modulation method such as 64QAM or 256QAM in which there is no on-axis data located on the I-axis or Q-axis, A, B, C, and D are determined in the same manner as above.

FIGS. 10 and 11 are diagrams showing a third operation example of the sign determination circuit. The third operation example is a case where the modulation signal is 8QAM. 8QAM is a transmission modulation method in which on-axis data located on the I axis or the Q axis of the IQ plane exists in the X polarization data and the Y polarization data. Two examples of FIG. 10 and FIG. 11 will be described regarding the association of on-axis data located on the I-axis or the Q-axis with quadrants. However, as described above, this correspondence is not limited to these two examples, and correspondence can be made to any quadrant as long as there is no overlap.

When the modulation method is 8QAM, the four outer shells of the eight data are mapped to the center of each of the first to fourth quadrants, but the remaining four inner shell data are mapped on the I-axis or Q-axis. mapped to Therefore, since on-axis data located on the I-axis or the Q-axis exists in step S0, all steps after step S1 in the flowcharts of FIGS. 6 and 7 are executed.

Power estimation is performed in step S1. The power of the X polarization data and the Y polarization data is calculated using the following formulas.
Power of X polarization data = Absolute value of X_I + Absolute value of X_Q Power of Y polarization data = Absolute value of Y_I + Absolute value of Y_Q

Regarding the four pieces of data in the outer shell, if the data at the time of reception is located approximately at the center of each quadrant, the absolute value of X_I and the absolute value of X_Q will be approximately the same value, and the absolute value of Y_I and the absolute value of Y_Q will be approximately the same value. Since the absolute values are also approximately the same, each power is calculated as approximately twice the absolute value of the I-axis coordinate value.

On the other hand, for the four inner shell data, either the absolute value of X_I or the absolute value of X_Q will be almost zero, and either the absolute value of Y_I or the absolute value of Y_Q will be almost zero . Therefore, the power is calculated only as the absolute value of the I-axis coordinate value or the absolute value of the Q-axis coordinate value, respectively.

Therefore, assuming that the coordinate value of the outer shell is approximately twice that of the inner shell, the power of the outer shell will be approximately four times that of the inner shell. The difference between their powers is relatively large and detectable. Therefore, by setting a threshold value between them, it is possible to relatively easily determine whether the received data is inner shell data or outer shell data.

In step S2, it is determined whether the received data is inner shell data or outer shell data. In the case of 8QAM, as described above, if the threshold is set to 2.5 times the I-axis coordinate value of the inner shell data, if the power estimated in step S1 is higher than the set threshold, the outer shell data will be lower. If so, it is determined to be inner shell data. Note that the threshold value is selected depending on the modulation method to a value that allows more reliable determination of the inner shell and the outer shell. By this determination, in the 8QAM example shown in FIG. 10, data located near the center of the four quadrants on the IQ plane is outer shell data, and on-axis data located on the I axis or Q axis is inner shell data. Identified as.

In step S3, the following processing is performed on the inner shell data. If the absolute value of X_I≧the absolute value of X_Q, and if X_I≧0, the sign determination unit If <0, the on-axis data is associated with the value indicating the second quadrant, and (A, B) = (-1, +1). If the absolute value of X_I<the absolute value of X_Q, and if X_Q≧0, the sign determination unit If it is 0, the on-axis data is associated with the value indicating the third quadrant, and (A, B) = (-1, -1). If the absolute value of Y_I≧the absolute value of Y_Q, the sign determination unit Y associates the on-axis data with the value indicating the fourth quadrant (A, B)=(+1, -1) if Y_I≧0, and sets Y_I to the value indicating the fourth quadrant. If <0, the on-axis data is associated with the value indicating the second quadrant, and (A, B) = (-1, +1). If the absolute value of Y_I<the absolute value of Y_Q, and if Y_Q≧0, the sign determination unit Y associates the on-axis data with the value indicating the first quadrant (A, B)=(+1, +1), and determines that Y_Q< If it is 0, the on-axis data is associated with the value indicating the third quadrant, and (A, B) = (-1, -1).

That is, regarding the inner shell data, positive data on the I-axis is associated with values indicating the fourth quadrant, negative data on the I-axis is associated with values indicating the second quadrant, and data on the Q-axis is associated with values indicating the second quadrant. The data on the positive side above is associated with the value indicating the first quadrant, and the data on the negative side on the Q axis is associated with the value indicating the third quadrant. This corresponds to the case where the phase of the signal point of each data is rotated clockwise by 45 degrees. The sign determination circuit 7 outputs the values of A, B, C, and D corresponding to each quadrant.

Regarding the inner shell data, the data on the positive side on the I axis is associated with the value indicating the first quadrant, the data on the negative side on the I axis is associated with the value indicating the third quadrant, and the data on the Q axis is associated with the value indicating the third quadrant. It is also possible to associate data on the positive side with values indicating the second quadrant, and associate data on the negative side on the Q axis with the fourth quadrant. This corresponds to the case where the phase of the signal point of each data is rotated counterclockwise by 45 degrees. This situation is shown in FIG.

Next, in step S4, the following processing is performed on the outer shell data. Regarding this processing, the example in FIG. 10 and the example in FIG. 11 are the same. In the case of X_I≧0, the sign determination unit 1). In the case of X_I<0, the sign determination unit , -1). If Y_I≧0, the sign determination unit Y determines (C, D)=(+1,+1) as the first quadrant if Y_Q≧0, and (C, D)=(+1,−1) as the fourth quadrant if Y_Q<0. 1). In the case of Y_I<0, the sign determination unit Y sets (C, D)=(-1, +1) as the second quadrant if Y_Q≧0, and (C, D)=(-1) as the third quadrant if Y_Q<0. , -1).

Regarding the 8QAM outer shell data shown in FIGS. 10 and 11, the outputs (A, B), (C, D) of the sign determination units The result is the same as that determined by hard decision.

FIG. 12 is a diagram showing a comparative example of 8QAM sign determination. In the comparative example, hard decisions are made on the I-axis and Q-axis for 8QAM signal points, like QPSK or 16QAM, regardless of the power level.

In the case of 8QAM, for the 8 signal points, for the 4 pieces of data in the outer shell, the hard decision results are as they are: a value indicating the first quadrant, a value indicating the second quadrant, a value indicating the third quadrant, and a value indicating the fourth quadrant. The values indicate the quadrants, and the corresponding codes (A, B) and (C, D) are output. However, for the four pieces of data in the inner shell, the sign determination based on the hard decision result is uncertain.

On the other hand, in the sign determination circuit 7 using the sign determination method shown in FIGS. 6 and 7, in the case of 8QAM shown in FIGS. 10 and 11, the hard decision results for the four data in the outer shell directly correspond to each quadrant. Furthermore, the four data in the inner shell are assigned to one of the four quadrants without uncertainty, and are determined as the code associated with that quadrant.

Therefore, the sign determination circuit 7 according to the present embodiment can reliably determine the sign without making an uncertain determination even in the case of a modulation method in which some data is mapped onto the I-axis or the Q-axis. . Also, they can be averaged. Even in correlation calculation using the output, correlation values can be stably obtained and accurate comparisons can be made. As a result, the polarization separation monitor 3 including the sign determination circuit 7 incorrectly detects equivalence convergence even in the case of a modulation method such as 8QAM in which some data is mapped onto the I-axis or the Q-axis. cases can be significantly reduced.

13 to 16 are diagrams showing the results of functional evaluation in 8QAM of the adaptive equalization circuit according to the embodiment. These are measurements of the correlation situation within the correlation calculation circuit 8 of the polarization separation monitor 3. However, in the adaptive equalization circuit 27, all the tap coefficients of the four FIR filters in the digital filter 1 are in a state where they have converged to values that allow polarization separation to be performed correctly. Since the calculation of the tap coefficients has correctly converged, it is originally desirable that the correlation value in the polarization separation monitor 3 be a low value. If the correlation value is high, even though the calculation of the tap coefficients has converged, it has not converged, which means that erroneous detection is being performed.

FIG. 13 shows the correlation value between the I-axis component X_I of the X-polarized wave data and the I-axis component Y_I of the Y-polarized wave data. FIG. 14 shows the correlation value between the I-axis component X_I of the X-polarization data and the Q-axis component Y_Q of the Y-polarization data. FIG. 15 shows the correlation value between the Q-axis component X_Q of the X-polarized wave data and the I-axis component Y_I of the Y-polarized wave data. FIG. 16 shows the correlation value between the Q-axis component X_Q of the X-polarized wave data and the Q-axis component Y_Q of the Y-polarized wave data. The vertical axis indicates the correlation value, where 1 indicates the maximum correlation and 0 indicates no correlation. The horizontal axis indicates time. In each graph, the upper graph (◆) indicates the output of the correlation calculation circuit of the conventional polarization separation monitor to which this embodiment is not applied. The lower graph (points ■) indicates the output of the correlation calculation circuit 8 of the polarization separation monitor 3 according to the present embodiment.

The output of the correlation calculation circuit of the conventional polarization separation monitor may show a maximum value of 0.55 in FIG. 14 and 0.65 or more in FIG. 16, and a large correlation value is detected. Assuming that the comparison value of the correlation value in the comparison circuit is 0.5, it can be seen that erroneous detection is performed with high frequency. If an erroneous detection is performed, the polarization separation monitor 3 determines that equivalence convergence has occurred and notifies the control circuit 4 of the determination. Based on this notification, the control circuit 4 instructs the filter tap coefficient updating circuit 2 to re-update. Even though the tap coefficients have converged to values that allow correct polarization separation, the updating work will be performed again. This causes the tap coefficients to fluctuate unnecessarily, making it impossible to perform stable adaptive equalization processing.

On the other hand, the output of the correlation calculation circuit 8 of the polarization separation monitor 3 according to this embodiment is 0.2 or less in all cases. If the comparison value of the correlation value in the comparison circuit 9 is 0.6, no erroneous detection will occur. As a result, the polarization separation monitor 3 does not notify the control circuit 4 of false equivalence convergence information. It is also possible to reduce unnecessary re-update instructions from the control circuit 4 to the filter tap coefficient update circuit 2. As a result, stable adaptive equalization processing can be continued without unnecessary fluctuations in the tap coefficients.

As shown above, the effectiveness of the polarization separation monitor 3 according to the present embodiment can be confirmed in actual evaluation as well. Furthermore, the adaptive equalization circuit 27 including the polarization separation monitor 3 according to the present embodiment is compatible with any modulation method and can perform stable polarization separation operation.

1 Digital filter, 2 Filter tap coefficient update circuit, 3 Polarization separation monitor, 4 Control circuit, 7 Sign determination circuit, 8 Correlation calculation circuit, 9 Comparison circuit, 25 AD converter, 26 Chromatic dispersion compensation circuit, 27 Adaptive equalization Circuit, 201 Receiving optical module

Claims (8)

a digital filter that inputs the polarization-separated first polarization signal and the second polarization signal and performs further polarization separation processing;
a filter tap coefficient update circuit that updates tap coefficients of the digital filter according to changes in polarization state;
a polarization separation monitor that compares a correlation value between the first polarization signal and the second polarization signal output from the digital filter with a predetermined value;
a control circuit that causes the filter tap coefficient updating circuit to re-execute initial convergence of the tap coefficients when the polarization separation monitor detects that the correlation value exceeds the predetermined value;
The polarization separation monitor includes a sign determination circuit that determines which quadrant of the four quadrants on the IQ plane the first polarization signal and the second polarization signal belong to, and a sign determination circuit of the sign determination circuit. comprising a correlation calculation circuit that calculates the correlation value from the determination result, and a comparison circuit that compares the correlation value with the predetermined value,
In the case of a transmission modulation method in which on-axis data located on the I-axis or Q-axis of the IQ plane is present in the first polarized signal and the second polarized signal, the sign determination circuit Of the on-axis data, data on the positive side of the I axis, data on the negative side of the I axis, data on the positive side of the Q axis, and data on the negative side of the Q axis are displayed on the IQ plane, respectively. An adaptive equalization circuit characterized in that the adaptive equalization circuit associates one value out of the four quadrants without overlap with each other. 2. The sign determination circuit extracts the on-axis data based on the magnitude of data power of the inputted first polarized signal and the second polarized signal. Adaptive equalization circuit. 3. The adaptive equalization according to claim 2, wherein the magnitude of the power is estimated based on the addition of the absolute value of the coordinate value on the I-axis and the absolute value of the coordinate value on the Q-axis of the data. circuit. Claims 1 to 4, wherein the values indicating the four quadrants on the IQ plane are (+1, +1), (-1, +1), (-1, -1), and (+1, -1). 3. The adaptive equalization circuit according to any one of 3. In the I-axis component and Q-axis component of each of the first polarization signal and the second polarization signal, if the absolute value of the I-axis component≧the absolute value of the Q-axis component, and if the I-axis component≧0, then the above associating the on-axis data with a value indicating the fourth quadrant; if the I-axis component < 0, associating the on-axis data with a value indicating the second quadrant;
If the absolute value of the I-axis component < the absolute value of the Q-axis component, if the Q-axis component ≧ 0, the on-axis data is associated with the value indicating the first quadrant, and if the Q-axis component < 0, the on-axis data is associated with the third quadrant. 4. The adaptive equalization circuit according to claim 1, wherein the adaptive equalization circuit is associated with a value indicating a quadrant. In the I-axis component and Q-axis component of each of the first polarization signal and the second polarization signal, if the absolute value of the I-axis component≧the absolute value of the Q-axis component, and if the I-axis component≧0, then the above associating the on-axis data with a value indicating the first quadrant; if the I-axis component < 0, associating the on-axis data with a value indicating the third quadrant;
If the absolute value of the I-axis component < the absolute value of the Q-axis component, if the Q-axis component ≧ 0, the on-axis data is associated with the value indicating the second quadrant, and if the Q-axis component < 0, the on-axis data is associated with the value indicating the second quadrant. 4. The adaptive equalization circuit according to claim 1, wherein the adaptive equalization circuit is associated with a value indicating a quadrant. a receiving optical module that converts the received optical signal into an electrical signal;
an AD converter that converts the electrical signal into a digital signal;
a chromatic dispersion compensation circuit that compensates for distortion due to chromatic dispersion of the digital signal;
4. A receiving device comprising: the adaptive equalization circuit according to claim 1, which performs equalization processing to compensate for distortion due to polarization fluctuation of the output signal of the chromatic dispersion compensation circuit. a step in which the digital filter inputs the polarization-separated first polarization signal and the second polarization signal and performs further polarization separation processing;
a step in which a filter tap coefficient updating circuit updates the tap coefficients of the digital filter according to changes in the polarization state;
A polarization separation monitor determines which quadrant of the four quadrants on the IQ plane the first polarization signal and the second polarization signal belong to, and separates the first polarization signal and the second polarization signal. calculating a correlation value with the polarized signal of No. 2 and comparing the correlation value with a predetermined value;
If the polarization separation monitor detects that the correlation value exceeds the predetermined value, the control circuit causes the filter tap coefficient updating circuit to re-execute initial convergence of the tap coefficients,
In the case of a transmission modulation method in which on-axis data located on the I axis or Q axis of the IQ plane exists in the first polarized signal and the second polarized signal, the polarization separation monitor Extract the data, and extract data on the positive side of the I-axis, negative data on the I-axis, positive data on the Q-axis, and data on the negative side of the Q-axis among the on-axis data on the IQ plane. An adaptive equalization method characterized by associating one value among the values indicating the above four quadrants without overlap.

Images